There is increasing evidence that platelet thrombi play an important role in the pathogenesis of acute myocardial infarction (AMI). We compared "spontaneous" platelet aggregation in whole blood in 17 non-diabetic and 12 diabetic subjects on admission with AMI. There was no significant difference in the fall in platelet count between the two groups, expressed as platelets remaining (75.2 +/- 7.9% vs 77.3 +/- 6.9% at 10 min, 66.6 +/- 8.9% vs 68.5 +/- 6.3% at 20 min, 63.5 +/- 8.2% vs 64.9 +/- 6.7% at 30 min and 59.4 +/- 10.3% vs 61.3 +/- 7.6% at 60 min). The rate of "spontaneous" aggregation was increased in subjects with evidence of heart failure on admission compared to those without (59.9 +/- 7.9% vs 66.2 +/- 6.6% at 30 min [p = 0.05] and 55.4 +/- 9.6% vs 63.1 +/- 7.7% at 60 min [p = 0.04]). There was no correlation between the fall in platelet count and admission plasma glucose, glycated haemoglobin or peak aspartate amino-transferase. The subjects studied on admission with AMI had greater rates of "spontaneous" aggregation than 8 subjects studied between 6 and 12 months after acute myocardial infarction (75.9 +/- 7.4% vs 85.8 +/- 5.4% at 10 min; p = 0.001 and 64.3 +/- 7.5% vs 75.0 +/- 7.8% at 30 min; p = 0.006) and compared to normal controls (90.7 +/- 4.4% at 10 min; p < 0.001 and 83.4 +/- 6.5 at 30 min; p < 0.001). This study provides evidence of increased "spontaneous" platelet aggregation in subjects admitted with acute myocardial infarction but no difference between diabetic and non-diabetic subjects was observed.